Currently, image pickup devices such as digital cameras and camera phones are widely used to take photographs. As known, the definition of the object taken by the image pickup device is largely effected by the focusing operation of the image pickup device. In order to achieve high image quality of the object, the focal length should be properly adjusted to focus on the object. In other words, the quality of the digital camera or the camera phone is highly dependent on the auto focus method applied to the digital camera.
Conventionally, there are several means for implementing focus value measurements. A common method for measuring focus value is for example a global search algorithm. Since the global search algorithm captures image in every lens movement step and determines the lens position corresponding to the maximum focus value, the global search algorithm may achieve the most correct search results among these lens position search algorithms. However, the global search algorithm needs too long search time and too many lens movement steps.
Since the lens needs to frequently move back and forth, the lens is readily suffer from mechanical backlash problem and the pot life thereof is shortened. Ideally, the curve plot illustrating the relation between lens search steps versus focus values is smooth. Whereas, if the focus values or sharpness values of the image are affected by noise, the curve plot illustrating the relation between lens search steps versus focus values becomes saw-toothed or rugged. The saw-toothed or rugged curve indicates erroneous judgment of peak position and erroneous focusing operation.
Therefore, there is a need of providing an auto focus method for achieving quick and correct focusing results.